Disaster Risk Analysis Part 1: The Importance of Including Rare Events

David A. Etkin, Aaida A. Mamuji, Lee Clarke

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Rare events or worst-case scenarios are often excluded from disaster risk analysis. Their inclusion can be very challenging, both from methodological and data availability perspectives. We argue that despite these challenges, not including worst-case scenarios in disaster risk analysis seriously underestimates total risk. It is well known that disaster data sets generally have fat tails. In this paper we analyze data for a number of disaster types in order to empirically examine the relative importance of the few most damaging events. The data show consistent fat-tail trends, which suggests that rare events are important to include in a disaster risk analysis given their percentage contributions to cumulative damage. An example of biased risk estimation is demonstrated by a case study of risk analysis of tanker spills off the western coast of Canada. Incorporating worst-case scenarios into disaster risk analysis both reduces the likelihood of developing fantasy planning documents, and has numerous benefits as evidenced by applications of foresight analysis in the public sector. A separate paper "Disaster Risk Analysis Part 2" explores how disaster risk analyses are operationalized in governmental emergency management organizations, and finds evidence of a systemic underestimation of risk.

Original languageEnglish (US)
Article number20170007
JournalJournal of Homeland Security and Emergency Management
Volume15
Issue number2
DOIs
StatePublished - Jan 1 2018

All Science Journal Classification (ASJC) codes

  • Business, Management and Accounting (miscellaneous)
  • Safety, Risk, Reliability and Quality
  • Safety Research

Keywords

  • disaster risk analysis
  • fat tails
  • policy
  • rare events
  • worst case scenario

Fingerprint Dive into the research topics of 'Disaster Risk Analysis Part 1: The Importance of Including Rare Events'. Together they form a unique fingerprint.

Cite this